6.1.1. Legal preface

This takes place within a multi-faceted legal framework. Compliance is a core measure of quality and good practice. For example, the management of disturbed/violent behaviour frequently involves interventions to which an individual does not – or cannot – consent. It is especially important that such interventions are in accordance with best practice.

Failure to act in accordance with the guideline may not only be a failure to act in accordance with best practice, but in some circumstances may have legal consequences. For example, any intervention required to manage disturbed behaviour must be a reasonable and proportionate response to the risk it seeks to address.

The service should ensure access to competent legal advice when required, in relation to the management of disturbed/violent behaviour.

The law provides the authority to respond to disturbed/violent behaviour in some circumstances, and it sets out considerations that are extremely important when service providers have to decide what action they may take. The contribution of the law to the management of disturbed/violent behaviour should be recognised as positive and facilitative.

All those involved in the short-term management of disturbed/violent behaviour in psychiatric in-patient settings and emergency departments should:

• be familiar with, in particular:
  • the relevant sections of the Mental Health Act 1983 and its current Code of Practice
  • the principles underlying the common law doctrine of ‘necessity’
  • the requirements of the relevant articles of the European Convention on Human Rights, including Article 2 (right to life) and Article 3 (the right to be free from torture or inhuman or degrading treatment or punishment), Article 5 (the right to liberty and security of person save in prescribed cases) and Article 8 (the right to respect for private and family life), and the principle of ‘proportionality’
  • the Health and Safety at Work Act 1974, which place duties on both employers and employees, and applies to the risk of violence from patients and the public
  • the Management of Health and Safety at Work Regulations 1992, which places specific duties on the employer to ensure suitable arrangements for the effective planning, organisation, control, maintenance and review of health and safety (these duties include ensuring that the risk assessments are undertaken and implemented)
• receive regular training on the legal aspects of the management of disturbed/violent behaviour
• ensure that a comprehensive record is made of any intervention necessary to manage an individual's disturbed/violent behaviour, including full documentation of the reason for any clinical decision
• ensure or contribute to ensuring that all aspects of the management of disturbed/violent behaviour are monitored on a regular basis, and that any consequential remedial action is drawn to the attention of those responsible for implementing it
• be aware of the obligations owed to a service user while their disturbed/violent behaviour is being managed, and of parallel obligations to other service users affected by the disturbed/violent behaviour, to members of staff, and to any visitors
• ensure or contribute to ensuring that any service user who has exhibited disturbed/violent behaviour should not be the subject of punitive action by those charged with providing them with care and treatment, and that where the disturbed/violent behaviour is thought to warrant criminal sanction, it is drawn to the attention of the proper authority.

6.2.1. Environment and alarm systems

Environmental factors are believed to be important determinants of disturbed/violent behaviour in psychiatric in-patient settings. A therapeutic environment is one that allows individuals to enjoy safety and security, privacy, dignity, choice and independence, without compromising the clinical objectives of the service. Comfort, noise control, light, colour and access to space will all have an impact on the well-being of both staff and service users. However, to date there has been very little research conducted to ascertain how the environment affects staff and service users of in-patient psychiatric settings.

The little existing research in this area has suggested that high traffic areas in in-patient units are the location of the largest number of assaults. Several studies have indicated that the highest proportion of assaults occur in either the day room/communal room or in the corridors (Carmel 1989; Coldwell and Naismith 1989; Lanza et al. 1993; Rosenthal et al. 1992), suggesting that assault frequency is related to either a chance encounter or that crowding (service user population density) is a significant factor. Studies of temporal variation show that most assaults occur during mealtimes and afternoons and increase in frequency until late evening (Carmel 1989; Lanza et al. 1993; Manfredini et al. 2001).

Recent national guidance documents have highlighted the need for in-patient psychiatric settings to not only be safe and secure for staff and service users, but further have recommended that the quality of design and finish should also be a prime consideration. Indeed, recent audit reports have indicated that many UK psychiatric in-patient facilities have failed to meet basic standards for a decent working or residential care environment and these wards are rated by staff, service users and visitors as noisy, hot, smelly and dirty (College Research Unit 2000; 2001).

Alarm systems are also an essential environmental safety feature in psychiatric in-patient settings. The report Violence and aggression to staff in health services, outlines three types of alarm system:

6.2.2. Prediction: antecedents, warning signs and risk assessment

While most service users in psychiatric in-patient settings are not disturbed/violent, a small minority place health care professional and other service users at serious risk of assault. Therefore the prediction of short-term disturbed/violent behaviour is not an outcome that is measured for its own sake, but is part of a risk management plan that works towards minimising disturbed/violent behaviour and aggression, allowing both service users and staff to feel safe. As a consequence, risk assessment must be seen as an essential intervention, possibly the single most important intervention, in the therapeutic management of disturbed/violent behaviour. Worryingly, a survey conducted in 1999 by the Standing Nursing and Midwifery Advisory Committee (SNMAC) found that risk management, which should logically follow from risk assessment, is poorly defined and practice is highly variable (SNMAC, 1999). Furthermore, they found that risk assessment was not regarded as an essential aspect of clinical practice (SNMAC, 1999). While nothing can ever be predicted with 100 per cent accuracy, prediction of short-term disturbed/violent behaviour and risk assessment is integral to the management of disturbed/violent behaviour in psychiatric in-patient healthcare settings. The recent UKCC – now the Nursing and Midwifery Council (NMC) – report stresses:

While it is absolutely clear that violence is often unpredictable, the use of comprehensive risk assessment materials, followed by a properly developed plan is an absolute pre-requisite for the recognition, prevention, and therapeutic management of violence (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p15, p22).

Much of the research pre-1995 (the cut-off point for the majority of the searches underlying the original RCPsych guideline) suggested that risk factors of short-term psychiatric in-patient disturbed/violent behaviour can be identified. Key risk factors appear to include a history of disturbed/violent behaviour, young age and number of admissions. However, Stein (1998) argues that the real challenge is not their identification, but in how they should be combined and weighted. He states:

The prediction of […] harm to others is a complex and unreliable synthesis of observed past behaviour (both inside and outside of hospital […]). The key predictors are well understood but there is much less agreement about how they should be weighed […]

Therefore the issue that faces mental health care professionals is how the best predictive validity can be attained. Three main approaches have been adopted:

• the clinical approach (‘first generation’)
• the actuarial approach (‘second generation’), which includes actuarial tools or checklists
• structured clinical judgement (‘third generation’).

Most of the literature prior to 1995 suggests that clinical judgement has poor positive predictive validity of around 33 per cent (Doyle and Dolan 2002). Therefore a ‘second generation’ of risk assessment studies adopted actuarial measures, in an attempt to raise the positive predictive validity of short-term psychiatric in-patient disturbed/violent behaviour. This actuarial approach depends on ‘assessors reaching judgements based on statistical information according to fixed and explicit rules’ (Doyle and Dolan 2002). Actuarial checklists have been created to enhance this process. Both the use of checklists and this general approach have been suggested to improve predictive validity (Doyle and Dolan 2002). However, there are noticeable disadvantages to this approach, in particular the tendency it generates to focus on static factors, such as history of disturbed/violent behaviour, demographic information and diagnosis, without taking individual service user needs into consideration.

Most recently, it has been suggested that prediction needs to be carefully slotted into a more holistic approach, which places emphasis on the empirical or static factors isolated by the actuarial approach, whilst combining it with clinicians' judgements. This ‘third generation’ approach, described by Doyle and Dolan (2002) as ‘structured clinical judgement’ has the advantage of placing emphasis on the service user as an individual and allowing risk to be seen as a moving rather than static entity, so that stage of disease, and any fluctuations in personal and environment factors are taken into consideration. Such an approach seems to mirror the objectives of the UKCC report, where it states that:

The assessment of risk is an essential part of the care and treatment of all patients. It is most important to stress that risk levels change. Therefore, […] the nature and level of risk should be subject to regular review (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p15, p22).

6.4.1. Service user perspectives

In recent years a great deal has been written within guidance material on the need to involve service users in their care. One of the guiding principles of the National Service Framework (NSF) on mental health is to involve service users and their carers in the planning and delivery of care (Mental Health National Service Framework 1999). This principle is echoed by the Department of Health, which argues that:

In order to create a genuinely patient-centred service several processes should be created to enable users to contribute to the design and delivery of care. The aim is to promote a non-judgemental, non-patronising, collaborative approach to care (Department of Health, Mental health policy implementation guide 2002, p14).

The UKCC has laid out a number of principles that they believe need to be met in order to fulfil such aims in relation to adult service users in psychiatric in-patient settings. It argues that:

• The prevention and management of disturbed/violent behaviour should primarily be viewed as an occupational problem, requiring a cohesive, multi-faceted organisational approach. The safety and homeliness of clinical areas, the quality of life in clinical areas and the nature of staff interventions with patients and the assessment of the needs of patients and their clinical management are at least as important in this regard as training in and use of any specific intervention strategies. The importance of these factors needs to be recognised and emphasised in training and practice (The recognition, prevention and therapeutic management of violence in mental health care 2002).
• Service users, their advocates, and their carers should be involved in reviews of policies, and their contribution to the planning and provision of training should be seen as essential. The inquiry into the death of David Bennett highlighted once more the need to consider race, culture, and ethnicity in all areas of policy, practice and training. The input by service users, advocates and carers noted above must be incorporated into these perspectives (The recognition, prevention and therapeutic management of violence in mental health care 2002).

6.4.2. Minority ethnic groups

The David Bennett Inquiry (2004) highlighted the importance of considering the needs of black and minority ethnic groups when managing disturbed/violent behaviour in the short-term. For the purpose of this guideline, the following definition of minority ethnic group has been adopted:

Minority ethnic group: A group which is numerically inferior to the rest of the population in a state, and in a non-dominant position, whose members possess ethnic, religious or linguistic characteristics which differ from those of the rest of the population and who, if only implicitly, maintain a sense of solidarity towards preserving their culture, traditions, religion or language. (F. Capotorti (1985) ‘Minorities’, in Bernhardt R et al. (editors) Encyclopedia of public international law. Amsterdam: Elsevier, vol.8, p.385.)

The importance of this area is widely recognised by health care professionals (Fernando 1998) and has recently been highlighted by a number of high profile inquiries. The most recent of which is the inquiry into the death of David ‘Rocky’ Bennett, an African Caribbean service user who died whilst being restrained on a secure unit.

The literature, around mental health and minority ethnic groups, highlights particular concerns relating to black and African Caribbean service users. For the purpose of this guideline the following definition of black, taken from They look after their own, don't they? (DH/Social Service Inspectorate 1998), has been adopted:

Black: refers to those members of the ethnic minority groups who are differentiated by their skin colour or physical appearance, and may therefore feel some solidarity with one another by reason of past or current experience, but who may have many different cultural traditions and values.

For this purpose of this guideline, the following definition of African Caribbean has been adopted:

Of or pertaining to both Africa and the Caribbean; used to designate the culture, way of life, etc or the characteristic style of music of those people of black African descent who are, or whose immediate forebears were, inhabitants of the Caribbean (West Indies) (Oxford English Dictionary Online).

It is maintained that black and particularly African Caribbean service users are over-represented within the mental health services in the UK, particularly in forensic settings. A variety of reasons have been advocated, including:

• prevalence of schizophrenia amongst African Caribbean service users (Ndegwa 2000)
• institutional racism (Sashidharan 2003; Department of Health 2005).

It is also suggested that recent shifts in Government policy have led to a more punitive approach within mental health services, particularly secure settings, and that young black African Caribbean men have been made to bear the burden of this altered approach (Fernando et al. 1998). Again it has been asserted that this burden reflects racial stereotyping that regards young African Caribbean men as ‘big, black and dangerous’ (Prins H, Big, black and dangerous? Report of the Committee of Inquiry into the death in Broadmoor hospital of Orville Blackwood and a Review of the deaths of two other Afro-Caribbean patients 1993). It is suggested that this stereotyping affects the treatment of African Caribbean service users within many mental health settings. (Littlewood and Lipsedge 1997).

As a result of the concerns relating to the treatment of African Caribbean service users, the review in this guideline has given particular attention to the short-term management of the disturbed/violent behaviour of African Caribbean service users in psychiatric in-patient settings. However, it has not done so to the exclusion of other ethnic groups.

6.4.3. Gender

As far as possible, gender needs must also be taken into consideration in the short-term management of disturbed/violent behaviour in psychiatric in-patient settings. For the purpose of this guideline the following definition of gender has been adopted:

Gender describes those characteristics of women and men that are socially determined, as opposed to ‘sex’, which is biologically determined. (Mainstreaming gender and women's mental health implementation guide 2003).

While general differences between men and women in terms of mental health have been recognised, (for example, women are more likely to self-harm and suffer from depression, and men more likely to experience earlier onset and more disabling courses of schizophrenia), a recent report by the Department of Health, The women and mental health strategy (2003) stresses that these differences should be used to inform our understanding of an individual, rather an obscure their individuality. A further report reinforced the message that women's mental health needs to be conducted in relation to an individual woman's experiences, beliefs and struggles, as well as her ethnic group, age and sexual preferences (Good practices in mental health 1996).

In terms of managing disturbed/violent behaviour in psychiatric in-patient settings, the main concern raised in The women and mental health strategy has been to identify gender specific needs, such as single-sex facilities, and to ensure that both male and female service users feel safe, listened to and involved in identifying and meeting gender related needs (Mainstreaming gender and women's mental health implementation guide 2003).

6.4.4. Other special concerns

This evidence review focuses specifically on disabilities, other than learning disabilities (excluded from this guideline), and aims to consider the effects of sensory impairment. It has been noted that service users with such sensory impairments are particularly vulnerable when managed using the interventions discussed in this guideline. One such example is the restraining of a deaf service user's hands, thereby preventing them from communicating.

Very little has been written on the needs of service users with a disability in relation to the short-term management of disturbed/violent behaviour in psychiatric in-patient settings.

6.5.1. De-escalation techniques

De-escalation (also referred to as ‘defusing’ or ‘talk-down’) involves the use of various psychosocial short-term techniques aimed at calming disruptive behaviour and preventing disturbed/violent behaviour from occurring. Every effort is made to avoid confrontation. This can include talking to the service user, often known as verbal de-escalation, moving service users to a less confrontational area, or making use of a specially designated space for de-escalation. Stevenson and Otto (1998) offer the following definition of verbal de-escalation:

What is verbal de-escalation? A nurse might describe it as “talking the patient down,” but it is actually a complex, interactive process in which a patient is redirected towards a calmer personal space.

There are competing theoretical approaches to de-escalation, including verbal de-escalation. Some approaches make use of communication theory (for example, Paterson and Leadbetter 1997), others of situational analysis (Rix 2001). All approaches emphasise the need to observe for signs and symptoms of anger and agitation, approaching the person in a calm controlled manner, giving choices and maintaining the service users dignity. Some approaches suggest mirroring the patient's mood. De-escalation techniques also emphasise the therapeutic use of the nurse's own personality and relationship with the person (use of self) as one method to interact therapeutically with the patient.

In all approaches to de-escalation, stress is laid on the need for training and self-awareness. For example, Rix (2001) comments:

Becoming competent at de-escalation is in itself a sophisticated activity requiring much more than just a theoretical understanding of aggression. It cannot be considered in purely academic terms. The practitioner must undertake a developmental process, resulting in highly evolved self-awareness enabling the skills of de-escalation to become instinctive.

However, a recent report notes that, despite the emphasis that is often placed on the importance of de-escalation, little research has been carried out into the effectiveness of any given approach, leaving nurses to contend with conflicting advice and theories:

Unfortunately, there has been little research conducted into the effectiveness of different approaches to de-escalation, or, for that matter, into the effectiveness of training in any given approach. As Paterson and Leadbetter (1999) note, there is no standard approach to de-escalation. At the same time, practitioners may be faced with contradictory advice provided in the context of differing theoretical explanations for the violent event (National Institute for Social Work Research Unit 2000, p24).

6.5.2. Observation

Although much of the research carried out on observation has been undertaken in relation to the management of suicide and self-harm, the UKCC report (Feb 2002), which focuses on the short-term management of disturbed/violent behaviour in psychiatric in-patient settings, argues that these principles form a good basis for the short-term management of disturbed/violent behaviour in psychiatric in-patient settings. The UKCC report (Feb 2002) recommends that the principles of observation found in Addressing acute concerns (1999) – a report that focuses on the management of suicide and self-harm – should be adopted nationwide.

Although the focus of the work on observation in Addressing acute concerns was on suicide and self-harm, there are obvious implications for the use of observation in recognising the possibility of violence occurring and for preventing interventions (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p24).

[…] observation (carried out as set out in Addressing acute concerns) should underpin all other strategies (The recognition, prevention and therapeutic management of violence in mental health care (2002) London: United Kingdom Central Council for Nursing, Midwifery and Mental Health Visiting, p24).

Addressing acute concerns defines observation as “‘regarding the patient attentively’ while minimising the extent to which they feel that they are under surveillance” (p2). The UKCC report (Feb 2002), regards observation as a ‘core nursing skill’ and ‘arguably a primary intervention in the recognition, prevention and therapeutic management of violence’ (The recognition, prevention and therapeutic management of violence in mental health care 2002) It suggests that observation must be a two-way relationship, established between a service user and a nurse, which is meaningful, grounded in trust, and therapeutic for the service user. This relationship is considered to be the basis on which risk assessment, violence management and a programme of supportive observation can then be undertaken (The recognition, prevention and therapeutic management of violence in mental health care 2002).

Addressing acute concerns outlines four levels of observation – general observation, intermittent observation, within eyesight, within arms length – which, with slight modification, have been adopted within this current guideline. Other reports and studies detail a variety of other terms and levels of observation. The UKCC report, The recognition, prevention and therapeutic management of violence in mental health care (2002) argues that there is a need for the terminology to be standardised, quoting the following passage from Addressing acute concerns:

Research on the nursing practice of observing patients who are at risk from self harm, or of causing harm to others, shows that there is no consistency in the definition of terms, principles or processes. In some trusts there is no written policy for observation. Trusts vary greatly in the indications for observation and in the personnel that are thought appropriate to perform it. Where policies and procedures do meet reasonable standards, they may not be implemented properly (Addressing acute concerns 1999, p15).

Whilst the UKCC report, The recognition, prevention and therapeutic management of violence in mental health care (2002) has stressed the value of observation, Addressing acute concerns suggests that both nurses and service users have found this a difficult intervention with many nurses considering it custodial and lacking in therapeutic value (Addressing acute concerns 1999).

6.6.1. Physical interventions

In the UK the physical intervention primarily used in the short-term management of disturbed/violent behaviour is manual holding, rather than the use of mechanical devices such as belts, body vests or handcuffs. These devices are rarely and only used in exceptional circumstances, usually within high security settings. Physical intervention is predominantly described in the literature as restraint. In this guideline, this terminology is avoided because of its association with particular techniques, associated with control and restraint (C&R) and its various modifications. C&R was originally developed in 1981 for prison staff and was taken up by the special hospitals in the mid 1980s. It is still widely used in the NHS, although modifications have been developed to make these techniques more appropriate to the therapeutic care of service users – for instance, C&R general services, which modifies uses of pain as a restraint technique (Wright 1999). Although still widely used, we believe that the association of the term ‘restraint’ with this approach is unhelpful, as a wide range of physical interventions are now currently employed, many very different from C&R or its modifications, such as MAPA, which makes use of therapeutic holding.

The use of pain compliance as a method of managing violent behaviour is controversial amongst health care professionals and service providers. Although practice currently continues in some services, the recommendations in this guideline severely restrict its use for rescue purposes only (see Section 8, para 1.8). For the purpose of this guideline, physical intervention is defined as:

A skilled hands-on method of physical restraint involving trained designated health care professionals to prevent individuals from harming themselves, endangering others or seriously compromising the therapeutic environment. Its purpose is to safely immobilise the individual concerned.

The current Code of Practice to the Mental Health Act 1983 states that physical intervention should be a last resort:

Physical restraint should be a last resort, only being used in an emergency where there appears to be a real possibility of significant harm if withheld. It must be of the minimum degree necessary to prevent harm and be reasonable in the circumstances. (18.10-18.11)

There appears to be a dearth of knowledge about current practice. The literature review undertaken for the UKCC report in 2002, found ‘no high quality studies that evaluated either the use of restraint or of seclusion in those with mental illness’ (The recognition, prevention and therapeutic management of violence in mental health care 2002) The rate of physical interventions per annum in the UK is currently unknown. At present the National Institute for Mental Health in England (NIMHE) is compiling a register of all the techniques used in the UK.

6.6.2. Seclusion

Seclusion is the formal placing of a service user in a specially designated room for the short-term management of disturbed/violent behaviour. While it is recognised that this intervention is unpopular with service users, it is sometimes the preferred course of action to prevent prolonged physical intervention where rapid tranquillisation is contra-indicated or when service users have indicated a preference for it in advance directives.

The RCPsych Council Report (41) argues that the definition of seclusion needs to be broad to allow for the seclusion room door being open, closed but unlocked or locked. Therefore, for the purpose of this guideline, the following definition of seclusion has been taken from the Code of Practice:

Seclusion is the supervised confinement of a patient in a room, which may be locked to protect others from significant harm. Its sole aim is to contain severely disturbed behaviour, which is likely to cause harm to others. Seclusion should be used as a last resort; for the shortest possible time. Seclusion should not be used as a punishment or threat; as part of a treatment programme; because of shortage of staff; where there is any risk of suicide or self-harm. Seclusion of an informal patient should be taken as an indicator of the need to consider formal detention.

Seclusion must be differentiated from asking a service user to go to a designated room for the purpose of calming down. The latter is a de-escalation technique and the seclusion room should not routinely be used for this purpose. Seclusion, if chosen, is not viewed as a therapeutic intervention. It simply allows for a period of calming in the service user and should always be managed in a designated room for seclusion, separating the service user from other service users and placing them in a positive milieu (Cashin 1996).

6.6.3. Rapid tranquillisation